The layer of powder or liquid is created by the stepwise sinking of the three dimensional part being built into a liquid vat or powder bed thereby presenting a new layer of untransformed material at the surface and retaining a planar focal plane for the laser at or near that surface. The application of material is distinct from the application of laser energy and the process is characterised in that the laser energy cannot address any layer lower or higher than the current layer in a manufacturing cycle. This is in distinction from methods that direct material into an energy beam and/or have multi-axis movement and can build or repair individual articles in multiple planes.
The absence of standardised nomenclature makes naming both this field and the alternatives difficult; however within the broad field of additive manufacturing this field is generally known as ‘powder bed additive manufacturing’ and its distinctions are well understood—there is a planar field on which material is applied generally and then separately selectively processed.
Alternative additive manufacturing methods outside this field generally apply the material selectively—typically adjacent the energy that will convert them to a solid and may be multi-axis. Examples are robot welders, Optomec LENS, Trumpf (5 axes) Laser Metal Deposition, Fraunhofer ILT Varioclad and other commercial systems of similar concept that apply material locally and on multiple axes, and adaptions of hot melt glue gun and inkjet print head concepts that are mechanically scanned across a flat bed to locally apply material on a single axis.
In the process of interest, the article is constructed layer by layer using photo polymerisation, sintering or melting locally to solidify material locally within the layer to form that part of the article which lies within that layer. This process takes place at the surface of a bed or vat of powder or liquid where there is a fixed focal plane of energy at or near that surface and the part being built is layer by layer lowered to present successive layers of unprocessed powder or liquid at the surface for selective solidification. Typically, and most conveniently, the solidification is induced by a scanned laser beam.
The laser beam is preferably scanned using mirrors to achieve a higher speed than can be achieved by mechanical scanning of the laser or material support. The use of small spot size (to provide sufficient power density to melt metal powder and provide fine detail) is then achievable because the scanning is sufficiently fast. The present state of the art laser beams have a beam quality factor of almost M2=1.
For a laser melting of metal elements or alloys, typically an Ytterbium doped fibre laser of 200 watts power and beam quality of M2<1.05 is used with a spot size of about 70 microns at the focal plane. The laser is directed using typical laser scan speeds of 1 m/sec to 2 m/sec (up to 7 m/sec as a selectable) through an F-theta lens towards a fixed focal plane at (or near) the layer of the part to be built.
Processes with layer thickness of 20 microns and a mean powder particle size of about 40 microns are known to produce high quality parts in steels, nickel alloys and cobalt chrome with mechanical properties similar to forged or cast and processes are in development for many other metal elements and alloys such as aluminium, titanium and gold alloys.
Although there are technically successful processes, these processes are inherently slow; they use a single laser spot to build each three dimensional part. There is now a demand for faster and cheaper processes and a scalable machine where parts can be built more quickly but without a loss of surface finish, accuracy and material quality.
In a metal powder process the build quality (including surface finish, lack of voids or cracks, crystallography and mechanical properties) is a function of power density, laser spot shape and size, layer thickness and many other process characteristics such as powder grain shape and size.
Increasing process speed cannot be achieved simply by increasing power and/or scan speed; all other parameters remaining the same and achieve the same build quality. A change in power density at the very least changes the process and above a certain energy level causes vapourisation of at least some of the metals in the alloy of the powder, rather than only melting the powder. On the other hand, if scan speed is increased, even with additional power, not enough time may be allowed for the laser energy to be absorbed and melt the powder; that is to say the dwell time is too short.
The parameters affecting these matters are set out in Industrial Application of Lasers by John F. Ready (ISBN0-12-583961-8) and in particular illustrated in FIG. 12-13, which are shown on page 333 of the second edition. This figure plots the regimes of laser power density and pulse durations suitable for material processing applications. It is demonstrated that a combination of increased power density and dwell time can rapidly lead to surface vapourisation and identifies an area in which the desirable small melted depth is achieved. This is located at a pulse duration of around 10−4 seconds at powers between 105 to 106 W/cm2. The pulse duration of this diagram is directly related to laser scan speeds in the process of this invention.
It therefore follows that it is necessary to keep the power density the same with a higher power laser, which means that at the focal plane a larger spot size is required and possibly thicker layers. The latter parameter is not particularly helpful because the laser energy is absorbed at the surface of the material and not in the bulk (as an electron beam is), however the powder bed may be approximately 50% dense so some laser energy penetrates and is reflected to a little beyond the top surface. Inevitably larger laser spot sizes will reduce the geometric precision of the parts to be made and the same is true for thicker layers.
Changing the power/spot size relationship and layer thickness also effectively means that a new process is being used. In industries, such as the aircraft industry, where parts are subjected to demanding certification requirements, such a change can mean that the whole certification process may have to be redone, which could outweigh any cost benefits achieved by the higher power process.
There is, therefore, a need for an additive layer manufacturing apparatus and process which can achieve higher manufacturing speeds, in a manner which does not just utilize increased power at a single laser spot and does not sacrifice build quality.